Turbine variable working condition state visualization method and system based on multi-parameter fusion

Abstract

The invention relates to a steam turbine variable working condition state visualization method and system based on multi-parameter fusion, and belongs to the technical field of steam turbine operation and maintaining.The method comprises the steps that multi-source heterogeneous parameter data of steam turbine operation are collected, and multi-source heterogeneous parameters comprise process parameters and mechanical parameters; processing multi-source heterogeneous parameter data through a data fusion algorithm, constructing a space-time incidence matrix, and quantifying a dynamic incidence relation between parameters; based on the space-time incidence matrix, state features are extracted, and state indexes representing the comprehensive operation state of the steam turbine are generated; mapping a dynamic association relationship between the state indexes and the parameters to a visual interface, wherein the visual interface comprises a graph structure used for representing the key parameters and association strength of the key parameters; the visual interface further comprises a historical state backtracking channel which supports real-time comparison between the current operation state and the historical state. The variable working condition state sensing capacity of the steam turbine is improved, the non-planned shutdown risk is reduced, and core technical support is provided for intelligent operation and maintenance.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of steam turbine operation and maintenance, and more particularly to a steam turbine variable working condition state visualization method and system based on multi-parameter fusion. BACKGROUND

[0002] As the core power equipment in the fields of thermal power generation, nuclear power and the like, the variable working condition state of a steam turbine directly affects the safety and economy of the unit. The existing monitoring technology has the following limitations:

[0003] Multi-source parameter isolated analysis: process parameters (such as steam pressure and temperature) and mechanical parameters (such as vibration and displacement) are independently monitored in different systems, and the data island leads to the lack of state correlation.

[0004] Dynamic correlation quantization difficulty: the traditional threshold alarm only focuses on the over-limit of a single parameter, and cannot capture the time-varying correlation between parameters, especially with a high false alarm rate under variable working conditions.

[0005] Non-intuitive state evaluation: the static index (such as the efficiency value) depending on the empirical formula is difficult to adapt to complex working conditions, and the two-dimensional panel cannot intuitively display the parameter network correlation.

[0006] Low efficiency of historical comparison: manual retrieval of historical data is required for fault diagnosis, and there is a lack of automatic matching mechanism for similar states, which delays the fault disposal opportunity. SUMMARY

[0007] To solve the above technical problems, the application provides a steam turbine variable working condition state visualization method and system based on multi-parameter fusion.

[0008] The technical scheme of the application is as follows:

[0009] The application provides a steam turbine variable working condition state visualization method based on multi-parameter fusion, which comprises the following steps:

[0010] Collecting multi-source heterogeneous parameter data of the steam turbine operation, the multi-source heterogeneous parameters including process parameters and mechanical parameters;

[0011] Processing the multi-source heterogeneous parameter data through a data fusion algorithm, constructing a space-time correlation matrix, and quantizing the dynamic correlation between parameters;

[0012] Based on the space-time correlation matrix, extracting state characteristics and generating state indexes representing the comprehensive state of the steam turbine operation;

[0013] Mapping the state indexes and the dynamic correlation between parameters to a visualization interface, the visualization interface including a graph structure for representing key parameters and their correlation strengths;

[0014] The visualization interface further includes a historical state backtracking channel, supporting real-time comparison between the current operation state and the historical state.

[0015] Preferably, the data fusion algorithm comprises:

[0016] Modeling the parameters and their relationships using a graph structure, where nodes represent parameters and edges represent the strength of the association between parameters;

[0017] Predicting the future trend of parameter data using a time series prediction model that learns the time-varying correlation between parameters based on historical parameter sequences.

[0018] Preferably, the state feature extraction step comprises:

[0019] Using a hybrid modeling method combining physical mechanism-based models and data-driven models, where the physical mechanism model is based on thermodynamic and kinetic equations of the steam turbine, and the data-driven model is used to learn the non-linear relationship between parameters.

[0020] Preferably, the graph structure of the visualization interface is a dynamic and interactive topology graph that can perform at least one of the following operations:

[0021] Intuitively perceive the real-time state of key parameters through changes in node size, color or shape;

[0022] Perceive changes in the strength of the association between parameters through changes in the width, color or transparency of the edges;

[0023] Click on a node or edge to drill down and view detailed data or historical trends of the corresponding underlying parameters.

[0024] Preferably, the historical state backtracking channel comprises:

[0025] Store snapshots of the historical operating state of the steam turbine, which contain key parameter values, state indicators and association data between parameters;

[0026] Use a time series similarity matching algorithm to automatically retrieve historical states with similar characteristics to the current state;

[0027] Display a side-by-side or superimposed comparison view of the current state and the selected historical state on the visualization interface.

[0028] Preferably, the method further comprises a pre-warning step:

[0029] Based on the spatio-temporal association matrix or the state indicators, construct a dynamic pre-warning model;

[0030] When an abnormal evolution pattern of the association between parameters or the state indicators deviates from the safe operating interval is detected, trigger a pre-warning signal;

[0031] The pre-warning signal is highlighted visually or audibly on the visualization interface and is linked to the associated parameter or state area.

[0032] Preferably, the pre-warning step further comprises:

[0033] Using the correlation analysis model, when the pre-warning is triggered, the potential failure propagation path is predicted.

[0034] The predicted failure propagation path and the possible affected parameters or components are displayed in the form of a visual path diagram on the visualization interface.

[0035] In another aspect, the present application also provides a steam turbine variable working condition state visualization system based on multi-parameter fusion, comprising:

[0036] A data acquisition module acquires multi-source heterogeneous parameter data of the steam turbine operation, and the multi-source heterogeneous parameters include process parameters and mechanical parameters.

[0037] A correlation construction module processes the multi-source heterogeneous parameter data through a data fusion algorithm, constructs a time-space correlation matrix, and quantifies the dynamic correlation between parameters.

[0038] A feature extraction module extracts state features and generates state indicators representing the comprehensive state of the steam turbine operation based on the time-space correlation matrix.

[0039] A visualization module maps the state indicators and the dynamic correlation between parameters to a visualization interface, and the visualization interface includes a graph structure for representing key parameters and their correlation strength.

[0040] A backtracking comparison module, the visualization interface further includes a historical state backtracking channel, supports real-time comparison of the current operating state with the historical state.

[0041] In another aspect, the present application also provides an electronic device having a computer program stored thereon, and the computer program is executed by a processor to implement the steam turbine variable working condition state visualization method based on multi-parameter fusion according to any embodiment of the present application.

[0042] In another aspect, the present application also provides a computer readable medium for storing one or more programs, and when the one or more programs are executed by one or more processors, the one or more processors implement the steam turbine variable working condition state visualization method based on multi-parameter fusion according to any embodiment of the present application.

[0043] The present application has the following beneficial effects:

[0044] 1. Multi-source heterogeneous data deep fusion, synchronous acquisition of process / mechanical parameters, construction of space-time correlation matrix, quantification of dynamic correlation strength (such as time-varying correlation coefficient of pressure-vibration), and solution to the problem of data island.

[0045] 2. Hybrid modeling improves state evaluation accuracy, fuses physical mechanism model (based on thermodynamic equation) and data-driven model (such as deep autoencoder), generates 0-100 comprehensive state index, and reduces error by more than 40% compared with traditional methods.

[0046] 3. Dynamic topology graph realizes intuitive perception, and the node size / color mapping parameter importance and real-time state can be intuitively perceived through graph structure visualization; the edge width / transparency dynamically reflects the correlation strength change (such as width proportional to correlation coefficient), and the drilling interaction is supported, and the historical trend can be traced by clicking the node / edge, and the operation response time is less than 0.5 seconds.

[0047] 4. Historical backtracking strengthens state comparison, automatically matches similar historical states based on DTW algorithm, and supports fault reproduction analysis through split screen comparison view, and improves diagnosis efficiency by 60%.

[0048] 5. Integration of early warning and fault propagation prediction, dynamic early warning model detects associated abnormalities or SVI deviation from the safety interval in real time, visualizes the path graph to locate the fault root (such as bearing wear → vibration intensification → efficiency decline), and predicts the fault propagation path 15 minutes in advance. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0050] It should be understood that the step numbers used herein are only for the convenience of description, and are not limited to the execution sequence of the steps.

[0051] It should be understood that the terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, unless otherwise clear from the context, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0052] The terms "include" and "contain" indicate the presence of the described features, whole, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, whole, steps, operations, elements, components and / or sets thereof.

[0053] The term "and / or" means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0054] Embodiment one:

[0055] To solve the problems in the prior art, the application provides a method for visualizing the variable working condition state of a steam turbine based on multi-parameter fusion, comprising the following steps:

[0056] Collecting multi-source heterogeneous parameter data of the steam turbine operation, the multi-source heterogeneous parameters including process parameters and mechanical parameters;

[0057] In this embodiment, through a sensor network (including temperature sensors, pressure sensors, vibration sensors, flow meters, etc.) installed in the steam turbine system, process parameters such as main steam pressure, exhaust temperature, flow rate and mechanical parameters such as bearing vibration amplitude, rotor eccentricity and axial displacement are collected in real time. The data is stored in a time sequence form, and the sampling frequency is dynamically adjusted according to the parameter type, such as a process parameter sampling frequency of 1 Hz and a mechanical parameter sampling frequency of 10 kHz.

[0058] Processing the multi-source heterogeneous parameter data through a data fusion algorithm, constructing a space-time correlation matrix, and quantifying the dynamic correlation between parameters;

[0059] Graph structure modeling: constructing a parameter relationship topology graph, wherein a node represents a single parameter (such as node i = steam pressure and node j = bearing vibration), and an edge represents the correlation strength between parameters (such as the correlation between pressure and vibration).

[0060] Time series prediction model: using an LSTM long short-term memory network to train the historical parameter sequence, learning the time-varying correlation between parameters, and predicting the parameter trend in the next 5 minutes (such as predicting the vibration amplitude change when the steam pressure abnormally rises).

[0061] Based on the graph structure model and the time series prediction result, a space-time correlation matrix M is generated t = [r ij ] n×n ; wherein r ij represents the dynamic correlation coefficient between parameters i and j within a time window t, which is calculated by fusing the Pearson correlation coefficient and Granger causality test.

[0062] Based on the space-time correlation matrix, state features are extracted and state indicators representing the comprehensive state of the steam turbine operation are generated;

[0063] As a preferred embodiment of the present embodiment, the state feature extraction step comprises: a hybrid modeling method combining a physical mechanism-based model and a data-driven model, wherein the physical mechanism model is based on the thermodynamic and kinetic equations of the steam turbine, and the data-driven model is used to learn the nonlinear relationship between parameters. Wherein:

[0064] Physical mechanism model: calculate the theoretical efficiency value based on the steam turbine thermodynamic equation such as the Fluergel formula;

[0065] Data-driven model: extract nonlinear features (such as abnormal correlation patterns) from the spatiotemporal correlation matrix using a DAE deep autoencoder.

[0066] State indicator generation: fuse the physical model output and the DAE feature to generate a comprehensive state indicator of 0-100, and the calculation formula is:

[0067] SVI = w1·Efficiency + w2·DAE Score ;

[0068] Wherein: SVI is the comprehensive state indicator value; w1, w2 are weight coefficients based on expert experience; Efficiency is the output of the physical model; DAE Score is the DAE output feature.

[0069] Map the state indicator and the dynamic correlation relationship between parameters to a visualization interface, which includes a graph structure for representing key parameters and their correlation strength;

[0070] As a preferred embodiment of the present embodiment, the graph structure of the visualization interface is a dynamic and interactive topology graph, which can perform at least one of the following operations:

[0071] Intuitively perceive the real-time state of key parameters through changes in node size, color or shape; such as calculating parameter importance by entropy weight method and adjusting node size according to the size of parameter importance value; color mapping real-time value state such as green representing normal operation and red representing over-limit operation.

[0072] Perceive the change of correlation strength between parameters through the change of edge width, color or transparency; the width is proportional to the correlation strength r ij ; the transparency increases as r ij decreases;

[0073] Click on the node or edge to drill down to view the corresponding underlying parameter detailed data or historical trend.

[0074] The visualization interface also includes a historical state backtracking channel to support real-time comparison between the current operating state and the historical state.

[0075] As a preferred embodiment of the present embodiment, the historical state backtracking channel comprises:

[0076] A snapshot of a historical operating state of the steam turbine is stored, the snapshot containing key parameter values, state indicators and data of correlations between parameters;

[0077] A time series similarity matching algorithm is used to automatically search for a historical state with similar characteristics to the current state;

[0078] A comparison view of the current state and the selected historical state is displayed side by side or superimposed on the visualization interface.

[0079] As a preferred embodiment of the present embodiment, the early warning step further comprises:

[0080] An associated relationship analysis model is used to predict a potential failure propagation path when early warning is triggered; in the present embodiment, the specific early warning triggering condition is that when the strength of a key edge in the correlation matrix mutates by more than 30% or the SVI value is less than 85 for three consecutive samplings, early warning is triggered, and the abnormal node / edge is marked with a flashing red border in the topology graph, and a buzzer alarm is triggered; ij >30% or the SVI value is less than 85 for three consecutive samplings, early warning is triggered, and the abnormal node / edge is marked with a flashing red border in the topology graph, and a buzzer alarm is triggered;

[0081] The predicted failure propagation path and the possible affected parameters or components are displayed in the form of a visual path graph on the visualization interface. Specifically, a failure propagation graph is constructed based on the correlation matrix, a directed graph search algorithm is used to locate the root failure point (such as bearing wear), and the affected upstream and downstream parameter nodes are marked in red.

[0082] Embodiment Two:

[0083] The present embodiment provides a steam turbine variable operating condition state visualization system based on multi-parameter fusion, comprising:

[0084] A data acquisition module acquires multi-source heterogeneous parameter data of the steam turbine operation, including process parameters and mechanical parameters;

[0085] An associated relationship construction module processes the multi-source heterogeneous parameter data through a data fusion algorithm, constructs a spatio-temporal correlation matrix, and quantifies the dynamic correlation between parameters;

[0086] A feature extraction module extracts state features based on the spatio-temporal correlation matrix and generates state indicators representing the comprehensive state of the steam turbine operation;

[0087] A visualization module maps the state indicators and the dynamic correlation between parameters to a visualization interface, which includes a graph structure representing key parameters and their correlation strengths;

[0088] A backtracking comparison module, the visualization interface further comprises a historical state backtracking channel, which supports real-time comparison of the current operating state with historical states.

[0089] Embodiment three:

[0090] The embodiment provides an electronic device, which stores a computer program, and the computer program is executed by a processor to implement a method for visualizing a variable working condition state of a steam turbine based on multi-parameter fusion according to any one of the embodiments of the application.

[0091] Embodiment four:

[0092] The embodiment provides a computer readable medium for storing one or more programs, and the one or more programs, when executed by one or more processors, cause the one or more processors to implement a method for visualizing a variable working condition state of a steam turbine based on multi-parameter fusion according to any one of the embodiments of the application.

[0093] In the embodiments of the present application, "at least one" means one or more, and "multiple" means two or more. The "and / or" describes the association relationship of the associated objects, which means that there can be three kinds of relationships, for example, A and / or B, which can represent the cases of A alone, A and B together, and B alone. Wherein A and B can be singular or plural. The character " / " generally represents that the associated objects before and after it are in an "or" relationship. "At least one of the following" and the like expressions mean any combination of these items, including any combination of single or multiple items. For example, at least one of a, b and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, wherein a, b, and c can be single or multiple.

[0094] Those of ordinary skill in the art can realize that each unit and algorithm step described in the embodiments disclosed herein can be implemented by electronic hardware, computer software and a combination of electronic hardware and computer software. Whether the functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0095] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0096] In several embodiments provided in the present application, any function, if realized in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in part, or parts of the technical solutions, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a series of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM for short), a random access memory (Random Access Memory, RAM for short), a magnetic disk or an optical disk, and various media that can store program codes. The above is only an embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the specification of the present application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A method for visualizing the variable operating conditions of a steam turbine based on multi-parameter fusion, characterized in that, Includes the following steps: Collect multi-source heterogeneous parameter data of steam turbine operation, including process parameters and mechanical parameters; Multi-source heterogeneous parameter data are processed by data fusion algorithms to construct a spatiotemporal correlation matrix and quantify the dynamic correlation between parameters. Based on the spatiotemporal correlation matrix, state features are extracted and state indicators characterizing the comprehensive operating state of the steam turbine are generated. The dynamic relationships between status indicators and parameters are mapped to a visualization interface, which includes a graph structure for representing key parameters and their correlation strength. The visualization interface also includes a historical status backtracking channel, which supports real-time comparison of the current running status with the historical status.

2. The method for visualizing the variable operating condition state of a steam turbine based on multi-parameter fusion according to claim 1, characterized in that: The data fusion algorithm includes: A graph structure is used to model the parameters and their relationships, where nodes represent parameters and edges represent the strength of the association between parameters; A time-series prediction model is used to predict the future trend of parameter data. The time-series prediction model learns the time-varying correlation between parameters based on historical parameter sequences.

3. The method for visualizing the variable operating condition state of a steam turbine based on multi-parameter fusion according to claim 1, characterized in that: The state feature extraction step includes: A hybrid modeling approach combining a physical mechanism-based model and a data-driven model is adopted, wherein the physical mechanism model is based on the thermodynamic and kinetic equations of a steam turbine, and the data-driven model is used to learn the nonlinear relationships between parameters.

4. The method for visualizing the variable operating condition of a steam turbine based on multi-parameter fusion according to claim 1, characterized in that: The visualization interface has a dynamic and interactive topological graph structure, which can perform at least one of the following operations: The real-time status of key parameters can be intuitively perceived through changes in node size, color, or shape. Changes in the strength of the correlation between parameters can be perceived by variations in the width, color, or transparency of the edges. Click on a node or edge to drill down and view detailed data or historical trends of the corresponding underlying parameters.

5. The method for visualizing the variable operating condition state of a steam turbine based on multi-parameter fusion according to claim 1, characterized in that: The historical state backtracking channel includes: Store historical operating status snapshots of the steam turbine, the snapshots containing key parameter values, status indicators, and data on the correlation between parameters; Using time series similarity matching algorithms, historical states with similar characteristics to the current state are automatically retrieved; The current state and the selected historical state are displayed side by side or overlaid on the visualization interface.

6. The method for visualizing the variable operating condition state of a steam turbine based on multi-parameter fusion according to claim 1, characterized in that: The method also Including early warning steps: Based on the spatiotemporal correlation matrix or the state indicators, a dynamic early warning model is constructed; When an abnormal evolution pattern is detected in the correlation between parameters or the status indicator deviates from the safe operating range, an early warning signal is triggered. The warning signal is highlighted on the visualization interface in a visual or audible manner and linked to the associated parameter or status area.

7. The method for visualizing the variable operating condition state of a steam turbine based on multi-parameter fusion according to claim 6, characterized in that: The early warning steps also include: By using a correlation analysis model, potential fault propagation paths can be predicted when an early warning is triggered; The predicted fault propagation path and the parameters or components that may be affected are displayed on the visualization interface in the form of a visual path diagram.

8. A visualization system for the variable operating condition of a steam turbine based on multi-parameter fusion, characterized in that, include: The data acquisition module collects multi-source heterogeneous parameter data of the steam turbine operation, including process parameters and mechanical parameters; The correlation construction module processes multi-source heterogeneous parameter data through data fusion algorithms, constructs a spatiotemporal correlation matrix, and quantifies the dynamic correlation between parameters. The feature extraction module extracts state features and generates state indicators that characterize the overall operating state of the steam turbine based on the spatiotemporal correlation matrix. The visualization module maps the dynamic relationships between status indicators and parameters to a visualization interface, which includes a graph structure for representing key parameters and their correlation strength. The backtracking and comparison module includes a historical state backtracking channel in the visualization interface, which supports real-time comparison of the current running state with the historical state.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for visualizing the variable operating conditions of a steam turbine based on multi-parameter fusion as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for visualizing the variable operating conditions of a steam turbine based on multi-parameter fusion as described in any one of claims 1 to 7.